It is important to express the specific growth rate of a fermentation reaction as a function of substrate and product concentration in developing bioprocess automation techniques such as modeling of the rector and controlling it via an advanced control scheme. Neural network model is applied to the identification from time series data since it is insensitive to the random data error and easy to implement. In this study, a neural network method has been developed to estimate the specific growth rate estimation from the time series state variable data and test the performance.